Image restoration by an iterative damped least-squares method with non-negativity constraint

In this paper we describe a new digital method for restoring linearly degraded images in the presence of noise. The restoration procedure is an iterative damped least-squares (DLS) algorithm that is based on the principle of damped least squares. This method acquires the advantage of tolerance to noise by incorporating additional constraints of non-negativity of the object and the adaptive technique. We also use a median filter in the proposed algorithm to suppress spiky noise. After discussing the convergence of the iterative DLS algorithm, we present some results of computer simulations that demonstrate the effectiveness of the proposed method.